INDEX

Explanations

words related to specific purposes or intentions

oai_token-act-pair · gpt-3.5-turbo

mentions of various types of purposes in legal, social, or operational contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Patriarch

-0.89

 Neighbor

-0.67

 Sisters

-0.67

 Payne

-0.65

Box

-0.64

 hill

-0.64

 Colossus

-0.63

 Ribbon

-0.63

 Grape

-0.62

Tup

-0.62

POSITIVE LOGITS

folios

1.02

chwitz

1.00

tesy

0.93

 guiActiveUn

0.87

oldown

0.87

 purposes

0.84

afe

0.82

ussia

0.82

tenance

0.81

ģĸ

0.80

Activations Density 0.010%

words related to specific purposes or intentions

mentions of various types of purposes in legal, social, or operational contexts

No Comments

No Known Activations